383
J. Chem. Eng. Data 1993,38, 383-385
Isobaric Vapor-Liquid Equilibria for Cyclopentane + Cyclohexene + 1,2-Dichloroethaneand the Three Constituent Binary Systems Dong-Syau Jan, Yao-Cheng Xie, and Fuan-Nan Tsai' Department of Chemical Engineering, National Cheng Kung University, Tainan, Taiwan 70101, Republic of China
Isobaric vapor-liquid equilibria at atmospheric pressure have been determined for the title ternary system and the three constituent binary systems by using a dynamic equilibrium still. The binary data were tested for thermodynamic consistency and were correlated by the Wilson, NRTL, and UNIQUAC equations. Predictions for the ternary system by these equations have been compared with the experimental data.
Introduction Vapor-liquid equilibria (VLE)are required for an engineering application such as in the design and operation of distillation equipment. Some sets of isobaric VLE have already been reported for the system cyclohexene + 1,2dichloroethane (1,2). However, to our knowledge, no VLE data for the ternary system cyclopentane + cyclohexene + 1,2-dichloroethaneand for the two other binaries are available. In this paper, we present the VLE data for this ternary system and the three constituent binary systems at the pressure of 101.3 kPa. For each binary system the activity coefficients are evaluated and are correlated with three liquid models. The performance of various liquid models to predict the ternary VLB data from the constituent binary data has also been investigated. Experimental Section Vapor-liquid equilibrium measurements were carried out in a dynamicequilibriumstill manufacturedby Fischer Laborund-Verfahrenstechnik (Germany). The still allows good mixing of the vapor and liquid phases and good separation of the phases once they reach equilibrium, and it prevents entrainment of liquid drops and partial condensation in the vapor phase. A detailed description of the apparatus and operating procedure has been reported elsewhere ( 3 , 4 ) . The boiling temperature T in the equilibrium still was measured with a mercury-in-glass thermometer (0.1 K divisions), calibrated against a standard thermometer. The pressure P was maintained constant to within fO.l kPa by an electronic regulator. The compositions of the liquid xi and vapor yi phases were determined by using a Shimadzu gas chromatograph, type GC-l4A, equipped with a flame-ionization detector. The chromatographiccolumn was 180cm long, ready packed with 0.1% SP-lo00 on 80/100 Carbopack, and operated isothermally at 413 K. Injection and detector temperatures were 413 and 473 K, respectively. Nitrogen gas was used as the carrier gas at a flow rate of 60 mL/min. The gas chromatograph was calibrated by using mixtures of known composition that were prepared gravimetrically. The uncertainty of composition measurements was estimated to be fO.OO1 mole fraction. The materials used for this experiment were purchased from Merck and had a minimum purity of 99 % . No further purification of the chemicals was attempted since gas chromatographic analysis failed to show any significant impurities. Some physical properties determined for the chemicals were compared with values from the literature as shown in Table I.
360
11111111111
360
+o
390
320' 0.0
"
0.2
"
'
I
0.4 0.8 X l h YI)
"
0.8
'
I
1.0
Figure I. T-x-y diagram for the cyclopentane (1) + cyclohexene(2) system at 101.3 kPa: (0) experimental data; (-1 NRTL equation. 360,
I
,
I
,
I
I
I
I
I
\
960
330
0.0
0.2
0.4 0.8 x 1 b r Yl)
0.8
1.0
Figure 2. T-x-y diagram for the cyclopentane (1) + 1,2dichloroethane (2) system at 101.3 kPa: (0)experimental data; (-) NRTL equation. Results and Discussion The VLE data of the three binaries are presented in Table I1 and are also plotted in Figures 1-3 together with literature data points (1,2). The system cyclohexene + l,2-dichloroethane forms an azeotrope, while the other systems do not. The activity coefficients yi in the liquid phase were calculated as In yi =
(B, - V?)(P - Pi", + .(E) + xiP; RT
* To whom correspondence should be addressed. 0021-9568/93/1738-0383$04.00/0 0 1993 American Chemical Society
384 Journal of Chemical and Engineering Data,
Vol.38,No. 3,1993
Table I. Physical Properties of Chemicals: Normal Boiling Points Tb,Densities p , and Refractive Indexes ztD TdK
a
compound
exptl
cyclopentane cyclohexene 1,2-dichloroethane
322.43 356.02 356.54
p(293.15K)/(g cm-3) exptl lit.0 0.7446 0.745 38 0.8105 0.810 96 1.2525 1.253 1
lit.a 322.412 356.129 356.633
n~(293.15K) exptl
lit."
1.4060 1.4462 1.4443
1.406 45 1.446 54 1.444 8
Reference 15.
Table 11. Isobaric VLE Data: Temperature T, Liquid-Phase q and Vapor-Phase y1 Mole Fractions, and Activity Coefficients yi for Binary Systems at 101.3 kPa
T/K
x1
Y1
353.50 352.70 350.10 348.90 345.80 344.20 343.00 340.55
0.0467 0.0602 0.1082 0.1310 0.1968 0.2336 0.2641 0.3263
349.75 346.65 342.95 339.75 336.35 335.05 333.75 355.55 354.50 353.70 353.05 352.70 352.20 351.90
Y1
Y1
Ya
0.1216 0.1523 0.2537 0.2968 0.4084 0.4625 0.5016 0.5790
YZ T/ K Xl Cyclopentane (1)+ Cyclohexene (2) 1.086 0.994 339.05 0.3672 1.077 0.996 336.85 0.4280 1.066 0.998 334.65 0.4974 1.063 1.001 332.15 0.5851 1.055 1.002 329.50 0.6809 1.051 1.003 328.25 0.7362 1.041 1.006 325.60 0.8498 1.040 1.004 324.15 0.9171
0.6206 0.6807 0.7353 0.7983 0.8566 0.8859 0.9401 0.9679
1.032 1.033 1.023 1.015 1.012 1.005 1.002 1.o00
1.011 1.012 1.027 1.032 1.046 1.052 1.066 1.090
0.0416 0.0664 0.1050 0.1514 0.2141 0.2450 0.2799
0.2234 0.3147 0.4278 0.5120 0.5976 0.6401 0.6606
Cyclopentane (1)+ 1,2-Dichloroethane (2) 2.464 1.011 331.65 0.3493 2.357 1.011 329.60 0.4366 2.236 0.994 328.35 0.5034 2.026 0.995 326.80 0.5960 1.839 0.996 325.30 0.6872 1.787 0.970 324.75 0.8251 1.676 1.004 323.20 0.8797
0.6882 0.7387 0.7661 0.8050 0.8384 0.8908 0.9192
1.487 1.357 1.267 1.179 1.115 1.004 1.019
1.101 1.147 1.220 1.324 1.499 1.850 2.111
0.0537 0.1152 0.1715 0.2308 0.2772 0.3705 0.4606
0.0891 0.1695 0.2367 0.2918 0.3349 0.4076 0.4733
Cyclohexene (1) 1,2-Dichloroethane (2) 1.687 1.004 351.85 0.5361 1.011 1.542 351.95 0.5853 1.480 1.017 352.15 0.6359 1.382 1.037 352.25 0.6410 1.334 1.047 353.05 0.7795 1.088 354.00 0.8736 1.233 1.140 354.95 0.9355 1.162
0.5257 0.5642 0.6060 0.6123 0.7261 0.8242 0.9041
1.111 1.088 1.070 1.069 1.018 1.003 0.999
1.195 1.225 1.253 1.247 1.399 1.522 1.580
Y1
+
358
Table 111. Correlation Parametera AIS,Ail, and ull and Average Deviation8 between Calculated and Experimental Boiling Temperatures AT and Vapor-Phase Mole Fractionm A m for the Binary Systems and Ternary System
356
5
354
352
Cyclopentane (1)+ Cyclohexene (2) Wilson 45.26 15.64 0.07 0.0007 NRTL ((212 = 0.3) 82.86 -21.86 0.07 0.0007 UNIQUAC -55.40 81.21 0.07 O.OOO6
,
3500.0
0.2
0.4
0.6
X l b r
Yi)
0.8
1.0
Figure 3. T-x-y diagram for the cyclohexene (1) + 1,2dichloroethane (2) system at 101.3 kPa: ( 0 )this work; (A) Mesnage and Marsan (1); ( 0 ) Mato et al. (2); (-) NRTL equation. where
6.. U m.. V - Bii - BJJ. . (2) The second virial coefficients Bij for both the pure components and the mixtures were determined according to Tsonopoulos's empiricalcorrelations(5). The molar volumes ViL of the saturated liquid were estimated by the modified Rackett equation (6, 7). The vapor pressures of the pure componentsPfwere obtained by use of the Antoine equation with the constants from Reid et al. (8). The thermodynamic consistency of the data was tested by using the Herington integral method (9)and the point-topoint method as described by Fredenslund et al. (10).The three binary systems studied here proved to be consistent in
Cyclopentane (1)+ 1,2-Dichloroethene (2) Wilson 259.47 477.92 0.35 0.0109 NRTL (a12 = 0.3) 300.07 391.51 0.37 0.0111 UNIQUAC 144.37 104.54 0.37 0.0111
Cyclohexene (1)+ 1,2-Dichloroethane(2) Wilson 131.18 294.86 0.05 0.0017 NRTL ((212 0.3) 55.26 360.64 0.05 0.0017 UNIQUAC 112.28 29.44 0.05 0.0017
Wilson
NRTL UNIQUAC
Cyclopentane (1)+ Cyclohexene (2) + 1,2-Dichloroethane (3) 0.90 0.0268 0.0134 0.0214 0.78 0.0120 0.0104 0.0116 0.79 0.0121 0.0104 0.0116
a The definitions of the Wilson, NRTL, and UNIQUAC equatione are given in ref 14.
both methods, presenting deviations below the limits established for each test. The experimental data were correlated by the Wilson (1I), NRTL (12),and UNIQUAC (13)equations. The definitions of the equations and the pure component parameters are given in ref 14. As recommended by Renon and Prausnitz (12), the mixture nonrandomnessparameter a12 in the NRTL equation was set as 0.3.
Journal of Chemical and Engineering Data, Vol. 38, No. 3, 1993 385 Table IV. Isobaric VLE Data: Temperature T, Liquid-Phase y and Vapor-Phase yi Mole Fractions, and Activity Coefficients 71 for the Cyclopentane (1) + Cyclohexene (2) + If-Dichloroethane (3) System at 101.3 Wa T/K xi x2 Y1 Y2 Y1 Y2 78 353.35 0.0130 0.9354 0.0350 0.8857 1.128 1.026 1.715 351.55 0.0340 0.9059 0.0926 0.8148 1.194 1.028 1.818 350.95 0.0245 0.0561 0.1268 0.0945 2.303 1.960 1.018 349.10 0.0664 0.8556 0.1812 0.7069 1.273 1.016 1.828 347.50 0.0574 0.1058 0.2300 0.1397 1.949 1.705 1.010 343.55 0.0979 0.1398 0.3659 0.1421 2.019 1.484 0.984 343.05 0.1137 0.3344 0.3268 0.2831 1.574 1.255 1.096 342.15 0.1317 0.3918 0.3517 0.3004 1.498 1.170 1.166 342.05 0.1267 0.1837 0.3936 0.1622 1.747 1.352 1.032 341.45 0.1542 0.4477 0.3674 0.3234 1.362 1.127 1.270 340.95 0.1811 0.5177 0.3907 0.3505 1.251 1.073 1.428 340.05 0.1798 0.3698 0.4231 0.2597 1.398 1.146 1.207 339.50 0.2184 0.4464 0.4453 0.3079 1.230 1.146 1.286 339.45 0.2130 0.4440 0.4406 0.2890 1.250 1.083 1.379 339.45 0.1847 0.2972 0.4463 0.2070 1.460 1.159 1.170 339.05 0.2190 0.4507 0.4585 0.2704 1.279 1.011 1.455 338.70 0.2138 0.4329 0.4624 0.2295 1.334 0.904 1.565 336.95 0.2980 0.3847 0.5273 0.2200 1.146 1.033 1.517 336.15 0.3097 0.4090 0.5628 0.2223 1.204 1.008 1.497 334.35 0.3488 0.3341 0.6205 0.1622 1.241 0.956 1.430 332.75 0.3882 0.2547 0.6655 0.1144 1.253 0.934 1.361 331.30 0.4545 0.2256 0.7038 0.0975 1.181 0.945 1.445 329.70 0.5610 0.2299 0.7664 0.0930 1.092 0.935 1.658 328.80 0.5626 0.1718 0.7687 0.0684 1.123 0.949 1.563 328.10 0.6165 0.1948 0.8028 0.0722 1.093 0.906 1.732 327.25 0.6483 0.1452 0.8096 0.0551 1.075 0.956 1.768 326.15 0.7114 0.1107 0.8451 0.0392 1.058 0.927 1.828 324.85 0.8103 0.0853 0.8978 0.0292 1.027 0.940 2.064 323.80 0.8840 0.0490 0.9331 0.0167 1.011 0.972 2.302
The values of binary parameters for each equation were determined with the simplexsearchprocedure. The following objective function was minimized 71,dd
OF=
%[(
- 71,expU)
71,exptl
+
i
(72,dmi
- 72,exptl):
1
Y2,exptl
(3) where N is the number of measurements. The binary parameters for the correlation equations are shown in Table 111,along with the averagedeviationsbetween the calculated and experimental boiling temperatures AT and vapor-phase mole fractions Ay1 where
N
(4) N
(5) The results indicate that all equations are equally suitable
to represent the data. Also, for the cyclohexene (1)+ 1,2-
dichloroethane (2) system the azeotropic temperature Tu and compositionnlfl predicted by the NRTL equation (351.9 K, 0.513) compare fairly well with those of Mesnage and Marsan (1) (352.3 K, 0.505) and Mato et al. (2) (352.3 K, 0.528). Table IV presents the VLE data for the ternary system cyclopentane + cyclohexene + 1,2-dichloroethane. The averagedeviationsin the calculatedboiling temperatures and vapor-phase mole fractions by using various liquid models coupled with the corresponding binary parameters are listed in Table 111. As observed,the results predicted by the NRTL and UNIQUAC equations are superior to those of the Wilson equation.
Literature Cited (1) Meanage, J.; Marsan, A. A. J. Chem. Eng. Data 1971,16,434. (2)Mato, F.;Gonzalez,G.;Arroyo, F. J. Chem. Eng. Data 1989,34,179. (3)Coon, J. E.; Auwaerter, J. E.; McLaughlin, E. Fluid Phuse Equilib. 1989.44.306. (4)Say& A?. A. Fluid Phase Equilib. 1991,63,341. (5) Tsonopouloe, C. AIChE J. 1974,20,263. (6) Spencer, C. F.; Danner, R. P. J. Chem. E m . Data 1972. 17,236. (7)Yamada, T.;Gunn, R. D. J. Chem. Eng. Data 1979,18,234. (8)Reid, R. C.; Prauenitz, J. M.; Sherwood, T. K. The Properties of Gases and Liquids, 3rd ed.; McGraw-Hilk New York, 1977. (9)Herington, E. F. G. J. Inst. Pet. 1961,37,457. (10)Frededund, A.; Gmehling, J.; Rasmwen, P. Vapor-Liquid Equilibria Using UNIFAC. A Group-Contribution Method; Elsevier: Amsterdam, 1977. (11) Wilson,G. M. J. Am. Chem. SOC.1964,86,127. (12)Renon, H.; Prauenitz, J. M. AIChE J. 1968,14,135. (13) Abrams, D. S.; Prausnitz, J. M. AIChE J. 1978,21,116. (14)Gmehling,J.;Onken,U. Vapor-LiquidEquilibriumData Collection; Chemietry Data Series; DECHEMA: Frankfurt, Germany, 1977; VOl. I, Part 1. (15) Riddick, J. A.; Bunger, W. B. Organic Solvents. Physical Properties
and Methods of Purification, 3rd ed.; Techniques of Chemistry; Wiley-Interscience: New York, 1970; Vol. 11.
Received for review June 23, 1992. Revised October 26, 1992. Accepted January 11,1993.
